Related content: Liver & Bile, Primovist, MRI

Lee SJ, Kim SY, Kim KW, Kim JH, Kim HJ, Lee MG, Yu ES. Hepatic Angiomyolipoma Versus Hepatocellular Carcinoma in the Noncirrhotic Liver on Gadoxetic Acid–Enhanced MRI: A Diagnostic Challenge. AJR Am J Roentgenol. 2016;207(3):562-70.

How to differentiate liver lesions with Gd-EOB-DTPA

Four imaging features may facilitate the differentiation between benign hepatic angiomyolipoma and hepatocellular carcinoma in non-cirrhotic livers.

Hepatic angiomyolipoma (AML) is a benign mesenchymal tumor and identified by MRI or CT through its fatty component. However, other hepatic lesions, amongst them hepatocellular carcinoma (HCC) may also have a fatty component and AML are often misdiagnosed as HCCs.

Jung Lee, University of Ulsan College of Medicine, Seoul, Korea, and colleagues performed a retrospective study to assess the imaging characteristics of hepatic AML on gadoxetate disodium Gd-EOB-DTPA (Primovist®)-enhanced MRI. They further aimed to identify helpful imaging features for differentiating AML from HCC in the non-cirrhotic liver.


Lee et al. included 18 patients with pathologically proven hepatic AMLs who had undergone Gd-EOB-DTPA-enhanced MRI (AML group). They additionally included 36 patients with non-cirrhotic liver with radiologically confirmed single HCC who had also undergone Gd-EOB-DTPA-enhanced MRI (HCC group).

Two abdominal radiologists in consensus reviewed the MR images of all patients. They were blinded to the lesion type. The radiologists assessed for:

  • tumor size, location and margin;
  • presence of fat;
  • tumor signal intensity (SI) on T1-and T2-weighted imaging and DWI;
  • the pattern of contrast enhancement;
  • presence of an early draining vein during the arterial phase;
  • presence of prominent intratumoral vessels and
  • presence of tumor capsules

Lee et al. further calculated tumor-to-liver contrast ratios of each lesion and sensitivity and specificity of each significant MRI result. They then performed statistical analysis to compare the results between the AML and HCC group.


Four analyzed imaging features differed significantly between AMLs and HCCs:

  1. isointensity on DWI (16.7% vs 0.0%; p = 0.03),
  2. washout in the portal-venous phase (61.1% vs 88.9%; p = 0.03),
  3. early draining veins (27.8% vs 2.8%; p = 0.01),
  4. intratumoral vessels (55.6% vs 22.2%; p = 0.03).

All of the identified characteristics showed low sensitivity (16.7–55.6%) and high specificity (77.8–100%) for AML diagnosis.

Lee et al. additionally observed a difference in the presence of fat. However, this was not statistically significant (50% vs. 30.6%, p= 0.23). Tumor-to-liver contrast ratios of AML and HCCs did not differ significantly, irrespective of the imaging phase.


Lee et al. identified four imaging characteristics that could facilitate the differentiation between AML and HCC on Gd-EOB-DPTA-enhanced MRI in the non-cirrhotic liver. However, the study was based on a small number of patients and there was a selection bias due to the study being retrospective. Larger and prospective studies are needed to confirm the results.